{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.958041958041958,
  "eval_steps": 9,
  "global_step": 70,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.027972027972027972,
      "grad_norm": 5.75,
      "learning_rate": 2e-05,
      "loss": 1.8618,
      "step": 1
    },
    {
      "epoch": 0.027972027972027972,
      "eval_loss": 1.856866478919983,
      "eval_runtime": 4.5304,
      "eval_samples_per_second": 3.532,
      "eval_steps_per_second": 1.766,
      "step": 1
    },
    {
      "epoch": 0.055944055944055944,
      "grad_norm": 5.5,
      "learning_rate": 4e-05,
      "loss": 1.8343,
      "step": 2
    },
    {
      "epoch": 0.08391608391608392,
      "grad_norm": 5.75,
      "learning_rate": 6e-05,
      "loss": 1.87,
      "step": 3
    },
    {
      "epoch": 0.11188811188811189,
      "grad_norm": 6.0625,
      "learning_rate": 8e-05,
      "loss": 1.6976,
      "step": 4
    },
    {
      "epoch": 0.13986013986013987,
      "grad_norm": 5.6875,
      "learning_rate": 0.0001,
      "loss": 1.0745,
      "step": 5
    },
    {
      "epoch": 0.16783216783216784,
      "grad_norm": 2.875,
      "learning_rate": 0.00012,
      "loss": 0.5387,
      "step": 6
    },
    {
      "epoch": 0.1958041958041958,
      "grad_norm": 2.90625,
      "learning_rate": 0.00014,
      "loss": 0.3055,
      "step": 7
    },
    {
      "epoch": 0.22377622377622378,
      "grad_norm": 0.51171875,
      "learning_rate": 0.00016,
      "loss": 0.0592,
      "step": 8
    },
    {
      "epoch": 0.2517482517482518,
      "grad_norm": 0.65625,
      "learning_rate": 0.00018,
      "loss": 0.0185,
      "step": 9
    },
    {
      "epoch": 0.2517482517482518,
      "eval_loss": 0.0595741793513298,
      "eval_runtime": 4.6321,
      "eval_samples_per_second": 3.454,
      "eval_steps_per_second": 1.727,
      "step": 9
    },
    {
      "epoch": 0.27972027972027974,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002,
      "loss": 0.1101,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 6.375,
      "learning_rate": 0.00019997080140801932,
      "loss": 0.1327,
      "step": 11
    },
    {
      "epoch": 0.3356643356643357,
      "grad_norm": 0.21484375,
      "learning_rate": 0.00019988322268323268,
      "loss": 0.0082,
      "step": 12
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.00019973731496914914,
      "loss": 0.0049,
      "step": 13
    },
    {
      "epoch": 0.3916083916083916,
      "grad_norm": 0.0693359375,
      "learning_rate": 0.00019953316347176488,
      "loss": 0.0024,
      "step": 14
    },
    {
      "epoch": 0.4195804195804196,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001992708874098054,
      "loss": 0.065,
      "step": 15
    },
    {
      "epoch": 0.44755244755244755,
      "grad_norm": 0.07568359375,
      "learning_rate": 0.0001989506399451051,
      "loss": 0.0029,
      "step": 16
    },
    {
      "epoch": 0.4755244755244755,
      "grad_norm": 0.134765625,
      "learning_rate": 0.0001985726080931651,
      "loss": 0.0053,
      "step": 17
    },
    {
      "epoch": 0.5034965034965035,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.00019813701261394136,
      "loss": 0.0056,
      "step": 18
    },
    {
      "epoch": 0.5034965034965035,
      "eval_loss": 0.02016551047563553,
      "eval_runtime": 4.6628,
      "eval_samples_per_second": 3.431,
      "eval_steps_per_second": 1.716,
      "step": 18
    },
    {
      "epoch": 0.5314685314685315,
      "grad_norm": 0.0791015625,
      "learning_rate": 0.00019764410788292722,
      "loss": 0.003,
      "step": 19
    },
    {
      "epoch": 0.5594405594405595,
      "grad_norm": 0.032470703125,
      "learning_rate": 0.0001970941817426052,
      "loss": 0.0011,
      "step": 20
    },
    {
      "epoch": 0.5874125874125874,
      "grad_norm": 0.447265625,
      "learning_rate": 0.00019648755533435518,
      "loss": 0.0445,
      "step": 21
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00019582458291091663,
      "loss": 0.0708,
      "step": 22
    },
    {
      "epoch": 0.6433566433566433,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0114,
      "step": 23
    },
    {
      "epoch": 0.6713286713286714,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001943311813257743,
      "loss": 0.0243,
      "step": 24
    },
    {
      "epoch": 0.6993006993006993,
      "grad_norm": 1.921875,
      "learning_rate": 0.0001935016242685415,
      "loss": 0.0655,
      "step": 25
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.515625,
      "learning_rate": 0.00019261746489577765,
      "loss": 0.0756,
      "step": 26
    },
    {
      "epoch": 0.7552447552447552,
      "grad_norm": 0.0201416015625,
      "learning_rate": 0.00019167921953165825,
      "loss": 0.0008,
      "step": 27
    },
    {
      "epoch": 0.7552447552447552,
      "eval_loss": 0.0004657781682908535,
      "eval_runtime": 4.6398,
      "eval_samples_per_second": 3.448,
      "eval_steps_per_second": 1.724,
      "step": 27
    },
    {
      "epoch": 0.7832167832167832,
      "grad_norm": 0.0186767578125,
      "learning_rate": 0.00019068743608505455,
      "loss": 0.0008,
      "step": 28
    },
    {
      "epoch": 0.8111888111888111,
      "grad_norm": 0.02392578125,
      "learning_rate": 0.00018964269372957038,
      "loss": 0.0008,
      "step": 29
    },
    {
      "epoch": 0.8391608391608392,
      "grad_norm": 0.134765625,
      "learning_rate": 0.000188545602565321,
      "loss": 0.0046,
      "step": 30
    },
    {
      "epoch": 0.8671328671328671,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.0001873968032626518,
      "loss": 0.0049,
      "step": 31
    },
    {
      "epoch": 0.8951048951048951,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.00018619696668800492,
      "loss": 0.0047,
      "step": 32
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.0240478515625,
      "learning_rate": 0.0001849467935121521,
      "loss": 0.0012,
      "step": 33
    },
    {
      "epoch": 0.951048951048951,
      "grad_norm": 0.032958984375,
      "learning_rate": 0.00018364701380102266,
      "loss": 0.0014,
      "step": 34
    },
    {
      "epoch": 0.9790209790209791,
      "grad_norm": 0.0247802734375,
      "learning_rate": 0.00018229838658936564,
      "loss": 0.001,
      "step": 35
    },
    {
      "epoch": 1.006993006993007,
      "grad_norm": 0.01458740234375,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0006,
      "step": 36
    },
    {
      "epoch": 1.006993006993007,
      "eval_loss": 0.00015493688988499343,
      "eval_runtime": 4.6565,
      "eval_samples_per_second": 3.436,
      "eval_steps_per_second": 1.718,
      "step": 36
    },
    {
      "epoch": 1.034965034965035,
      "grad_norm": 0.00191497802734375,
      "learning_rate": 0.00017945776797137543,
      "loss": 0.0001,
      "step": 37
    },
    {
      "epoch": 1.062937062937063,
      "grad_norm": 0.00726318359375,
      "learning_rate": 0.00017796743540632223,
      "loss": 0.0003,
      "step": 38
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.002288818359375,
      "learning_rate": 0.00017643157205458483,
      "loss": 0.0001,
      "step": 39
    },
    {
      "epoch": 1.118881118881119,
      "grad_norm": 0.0029144287109375,
      "learning_rate": 0.00017485107481711012,
      "loss": 0.0001,
      "step": 40
    },
    {
      "epoch": 1.1468531468531469,
      "grad_norm": 0.0009918212890625,
      "learning_rate": 0.00017322686665977737,
      "loss": 0.0001,
      "step": 41
    },
    {
      "epoch": 1.1748251748251748,
      "grad_norm": 0.000743865966796875,
      "learning_rate": 0.00017155989607441213,
      "loss": 0.0,
      "step": 42
    },
    {
      "epoch": 1.2027972027972027,
      "grad_norm": 0.001007080078125,
      "learning_rate": 0.00016985113652489374,
      "loss": 0.0001,
      "step": 43
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.002471923828125,
      "learning_rate": 0.00016810158587867973,
      "loss": 0.0001,
      "step": 44
    },
    {
      "epoch": 1.2587412587412588,
      "grad_norm": 0.00103759765625,
      "learning_rate": 0.00016631226582407952,
      "loss": 0.0001,
      "step": 45
    },
    {
      "epoch": 1.2587412587412588,
      "eval_loss": 2.9456097763613798e-05,
      "eval_runtime": 4.6586,
      "eval_samples_per_second": 3.434,
      "eval_steps_per_second": 1.717,
      "step": 45
    },
    {
      "epoch": 1.2867132867132867,
      "grad_norm": 0.000637054443359375,
      "learning_rate": 0.00016448422127361706,
      "loss": 0.0,
      "step": 46
    },
    {
      "epoch": 1.3146853146853146,
      "grad_norm": 0.000431060791015625,
      "learning_rate": 0.00016261851975383137,
      "loss": 0.0,
      "step": 47
    },
    {
      "epoch": 1.3426573426573427,
      "grad_norm": 0.0020904541015625,
      "learning_rate": 0.00016071625078187114,
      "loss": 0.0001,
      "step": 48
    },
    {
      "epoch": 1.3706293706293706,
      "grad_norm": 0.000568389892578125,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.0,
      "step": 49
    },
    {
      "epoch": 1.3986013986013985,
      "grad_norm": 0.00135040283203125,
      "learning_rate": 0.00015680647467311557,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 1.4265734265734267,
      "grad_norm": 0.000621795654296875,
      "learning_rate": 0.00015480125073546704,
      "loss": 0.0,
      "step": 51
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0001527640244106133,
      "loss": 0.0953,
      "step": 52
    },
    {
      "epoch": 1.4825174825174825,
      "grad_norm": 0.005279541015625,
      "learning_rate": 0.00015069598538135906,
      "loss": 0.0002,
      "step": 53
    },
    {
      "epoch": 1.5104895104895104,
      "grad_norm": 0.0101318359375,
      "learning_rate": 0.0001485983413242606,
      "loss": 0.0004,
      "step": 54
    },
    {
      "epoch": 1.5104895104895104,
      "eval_loss": 0.0003907401696778834,
      "eval_runtime": 4.6412,
      "eval_samples_per_second": 3.447,
      "eval_steps_per_second": 1.724,
      "step": 54
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.007568359375,
      "learning_rate": 0.00014647231720437686,
      "loss": 0.0003,
      "step": 55
    },
    {
      "epoch": 1.5664335664335665,
      "grad_norm": 0.0167236328125,
      "learning_rate": 0.00014431915455992414,
      "loss": 0.0007,
      "step": 56
    },
    {
      "epoch": 1.5944055944055944,
      "grad_norm": 0.0703125,
      "learning_rate": 0.00014214011077725292,
      "loss": 0.0033,
      "step": 57
    },
    {
      "epoch": 1.6223776223776225,
      "grad_norm": 0.046142578125,
      "learning_rate": 0.00013993645835656953,
      "loss": 0.0022,
      "step": 58
    },
    {
      "epoch": 1.6503496503496504,
      "grad_norm": 0.32421875,
      "learning_rate": 0.00013770948416883205,
      "loss": 0.016,
      "step": 59
    },
    {
      "epoch": 1.6783216783216783,
      "grad_norm": 0.12060546875,
      "learning_rate": 0.00013546048870425356,
      "loss": 0.0058,
      "step": 60
    },
    {
      "epoch": 1.7062937062937062,
      "grad_norm": 0.0341796875,
      "learning_rate": 0.00013319078531285285,
      "loss": 0.0019,
      "step": 61
    },
    {
      "epoch": 1.7342657342657342,
      "grad_norm": 0.0166015625,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.001,
      "step": 62
    },
    {
      "epoch": 1.762237762237762,
      "grad_norm": 0.01190185546875,
      "learning_rate": 0.00012859456783986893,
      "loss": 0.0007,
      "step": 63
    },
    {
      "epoch": 1.762237762237762,
      "eval_loss": 0.00017181402654387057,
      "eval_runtime": 4.6193,
      "eval_samples_per_second": 3.464,
      "eval_steps_per_second": 1.732,
      "step": 63
    },
    {
      "epoch": 1.7902097902097902,
      "grad_norm": 0.005401611328125,
      "learning_rate": 0.0001262707378198587,
      "loss": 0.0003,
      "step": 64
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.00157928466796875,
      "learning_rate": 0.0001239315664287558,
      "loss": 0.0001,
      "step": 65
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.0022125244140625,
      "learning_rate": 0.00012157841967678063,
      "loss": 0.0002,
      "step": 66
    },
    {
      "epoch": 1.8741258741258742,
      "grad_norm": 0.00225830078125,
      "learning_rate": 0.00011921267173537086,
      "loss": 0.0001,
      "step": 67
    },
    {
      "epoch": 1.902097902097902,
      "grad_norm": 0.00179290771484375,
      "learning_rate": 0.00011683570413470383,
      "loss": 0.0001,
      "step": 68
    },
    {
      "epoch": 1.93006993006993,
      "grad_norm": 0.00188446044921875,
      "learning_rate": 0.00011444890495692213,
      "loss": 0.0001,
      "step": 69
    },
    {
      "epoch": 1.958041958041958,
      "grad_norm": 0.0012054443359375,
      "learning_rate": 0.0001120536680255323,
      "loss": 0.0001,
      "step": 70
    }
  ],
  "logging_steps": 1,
  "max_steps": 140,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 35,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0425496069000397e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}